RiceVision

BSc (Hons) Computer Science | Software Development Group Project

Rakindu Niwunhella

Chethina Kovida Fernando

Binada Matara Arachchige

Sanila Wijesekara

Sithuli Basnayake

Pavithma Fernando

The RiceVision system is a comprehensive agricultural monitoring and yield prediction platform designed to support data-driven decision making in rice cultivation. The system integrates satellite-based remote sensing, machine learning models and cloud computing technologies to analyse crop health and forecast yields efficiently. An automated data processing pipeline was developed using cloud services to handle data collection, preprocessing, and model execution. Satellite data obtained through Google Earth Engine and additional sources are processed and stored using scalable cloud infrastructure, enabling real-time access to insights. The backend system, implemented using FastAPI, manages data processing, API services and integration with a cloud-based database, ensuring efficient and secure data handling. On the frontend, an interactive dashboard was developed to visualise key agricultural metrics, including crop health indicators, yield predictions, and regional analytics. Additional features, such as notification systems, enhance user engagement by providing timely alerts and updates. The system was deployed using modern cloud platforms, ensuring reliability, scalability, and continuous availability. Overall, RiceVision demonstrates how the integration of remote sensing, machine learning, and cloud technologies can improve agricultural productivity, optimise resource management, and support sustainable farming practices.